Real-Time Payments Stress-Test Bank Infrastructure

Real-time payments are becoming a baseline expectation in a digital economy where consumers and businesses want money to move instantly, predictably and transparently, regardless of channel or rail.

For issuers, that shift is forcing a rethink of technology and strategy. According to Marjorie Tart, director of product management at i2c, the divide between success and struggle often comes down to how institutions frame real-time payments from the start.

From Technical Requirement to Strategic Imperative

When issuers approach real-time payments purely as a technical exercise, the focus narrows to rail connectivity. That mindset limits the upside, Tart said in an interview with PYMNTSTV.

“When I talk to issuers, the difference is usually whether they’re thinking about instant payments through a technical lens or a business lens,” she said. “Institutions in the first category are really just focused on simply connecting to a rail.”

Issuers that take the broader view treat real time as a growth lever, she said. They rethink customer journeys, funding flows and operational models rather than bolting instant payments onto batch systems designed decades ago.

“What we see at i2c is that issuers who succeed are the ones that treat real time as an opportunity to redesign their funding flows, reduce support costs, but also speed up production and innovation,” Tart said.

Use Cases That Expose Old Assumptions

The most compelling real-time use cases are often the ones that reveal the weaknesses of legacy infrastructure. Instant disbursements, account funding and account-to-account transfers are early pressure points, Tart said.

“In the old world, batch error systems introduced delays, breakage and manual intervention,” she said. “These use cases really depend on real-time availability of funds.”

Marketplace and gig payouts have added urgency. Speed has become a competitive differentiator, not a nice-to-have. But enabling these use cases requires more than access to a payment rail.

“You’ve got to have real-time posting, configurable workflows, routing logic, fraud controls and exception management,” Tart said.

When all that works seamlessly behind the scenes, issuers can monetize real-time capability more effectively, she said.

What Customers Want Is Clarity

While rails and protocols matter internally, customers care about outcomes. Expectations are consistent across consumer and business use cases, Tart said.

“They want instant, they want a clear status update, and they want a predictable experience regardless of what the channel is,” she said.

Problems arise when issuers attempt to deliver those experiences through disconnected systems. Card push payments, FedNow® Service transactions and ACH workflows often live in separate silos, creating inconsistent outcomes.

That’s where issuers fall short.

“They try to deliver using disconnected systems and then expect a consistent experience,” she said.

Different Issuer Approaches, Different Outcomes

Issuers also diverge in how they organize for real time internally. Some treat each rail as a standalone operation, building separate logic, controls and workflows for each integration.

That fragmentation makes it harder to scale and harder to govern. In contrast, issuers that connect once through a unified orchestration layer avoid rebuilding operations every time a new rail or use case emerges.

The Cost of Fragmented Connectivity

One of the biggest pain points issuers encounter is the operational burden of connecting separately to each payment rail. Each integration brings its own funding logic, limits, exceptions and reconciliation processes.

“A unified orchestration means issuers connect once,” Tart said.

Behind the scenes, i2c handles connectivity to FedNow, real-time payments and ACH through single APIs and a single set of rules.

Without that approach, institutions risk recreating the same complexity that real-time payments were supposed to eliminate.

Legacy Infrastructure Meets Always-On Reality

Batch-based cores struggle in an environment where balances, limits and liquidity must be known continuously. Issuers must embrace real-time ledgering and posting as defaults rather than add-ons, Tart said.

Funding models, limit management and exception handling all change in an always-on environment. Manual processes that once worked overnight no longer meet customer expectations measured in seconds.

Unified Orchestration and Continuous Intelligence

Real-time payments also compress decision windows for fraud, anti-money laundering and authorization. Post-transaction monitoring is no longer sufficient.

“You need to have continuous intelligence,” Tart said. “Behavioral analytics, velocity controls and dynamic limits that can adjust automatically.”

Unified risk scoring across rails is critical. An anomaly detected in one channel should inform decisions in another. Centralized governance ensures that fraud, compliance, treasury and product teams operate from the same data and rules.

Real Time as a Foundation, Not a Feature

As volumes grow, issuers are learning that real-time payments are not just about speed. They are about resilience, transparency and confidence.

“When you put these capabilities together, issuers can innovate faster and operate with much greater confidence,” Tart said.